1. Field
Embodiments of the invention relate to the field of networking; and more specifically, to fast reroute in networking.
2. Background Information
Network failures occasionally occur in Internet Protocol (IP) networks. By way of example, the network failures may be due to a failure of one or more routers or other network devices, a failure of one or more links, a failure of one or more links within a shared risk link group, or the like, or a combination thereof.
A significant problem with such network failures is the data loss that may occur if actions are not taken to restore the network failure. Routing protocols typically used in such IP networks generally have relatively long convergence times following a network topology change due to such network failures. Representatively, the convergence times of the routing protocols may be on the order of hundreds of milliseconds, or even longer. Such relatively long convergence times may cause significant data loss to occur if other actions are not taken to restore the network failure.
IP fast reroute technologies are commonly employed in IP networks in order to help reduce traffic loss following such network failures. By way of example, IP fast reroute may involve pre-computing or otherwise pre-determining alternate or backup next hops for corresponding primary next hops. The forwarding planes of routers may be pre-provisioned with both the primary next hop used for normal traffic forwarding before the network failure, as well as the backup next hop to be used for traffic forwarding after the network failure. After detecting a failure associated with the primary next hop, the forwarding planes and/or routers may rapidly switch from the primary next hop to the backup next hop, and may begin forwarding network traffic using the backup next hop. The switch from the primary next hop to the backup next hop generally occurs relatively rapidly. Representatively, the switch from the primary to backup next hop may occur within a period of time that is on the order of tens of milliseconds, or even less. Such times are typically significantly less than the times needed for the routing protocols to gather updated network information re-converge. The backup next hop may provide a temporary backup or alternate path on which the network traffic may be forwarded while the routing protocols re-converge. Advantageously, such use of IP fast reroute may help to significantly reduce data loss following network failures.
Various versions of IP fast reroute are known in the arts. RFC 5286, entitled Basic Specification for IP Fast Reroute: Loop-Free Alternates, published September 2008, discloses further details of one known approach to employ IP fast reroute in a domain or routing area.